A data science team is tasked with deploying a pre-built anomaly detection model in Snowflake to identify fraudulent transactions. They need to use Snowflake ML functions and a Snowflake Native App (that houses the model) to achieve this. The Snowflake Native App is installed and available. The transaction data is stored in a table called 'TRANSACTIONS. Which of the following steps are essential to successfully deploy and use this pre-built model within a User Defined Function (UDF) for real-time scoring, assuming the app provides a function named 'ANOMALY SCORE?

• A. Ensure the 'TRANSACTIONS' table is shared with the Snowflake Native App's container so the model can directly access the data.
• B. Create an external function in API Integration instead of UDF.
• C. Grant the USAGE privilege on the Snowflake Native App to the role executing the UDF. This ensures the UDF can access the app's functionality.
• D. Train the pre-built anomaly detection model using Snowflake's ML functions (e.g., 'CREATE MODELS) with the 'TRANSACTIONS' data before creating the UDE
• E. Create a UDF that calls the 'ANOMALY _ SCORE function provided by the Snowflake Native App, passing the relevant transaction features as arguments.

You are analyzing sales data in Snowflake using Snowpark to identify seasonality. You have a table named 'SALES DATA with columns 'SALE DATE (TIMESTAMP NTZ) and 'AMOUNT (NUMBER). You want to calculate the rolling average sales for each week over a period of 12 weeks using a Snowpark DataFrame. Which of the following Snowpark code snippets correctly implements this calculation?

• A.
  
• B.
  
• C.
  
• D.
  
• E.
  

You've built a complex machine learning model using scikit-learn and deployed it as a Python UDF in Snowflake. The UDF takes a JSON string as input, containing several numerical features, and returns a predicted probability However, you observe significant performance issues, particularly when processing large batches of data'. Which of the following approaches would be MOST effective in optimizing the performance of this UDF in Snowflake?

• A. Increase the warehouse size to improve the overall compute resources available for UDF execution.
• B. Serialize the scikit-learn model using 'joblib' instead of 'pickle' for potentially faster deserialization within the UDF.
• C. Pre-process the input data outside of the UDF using SQL transformations, reducing the amount of data passed to the UDF and simplifying the Python code.
• D. Use Snowflake's vectorized UDF feature to process data in micro-batches, minimizing the overhead of repeated Python interpreter initialization.
• E. Rewrite the UDF in Java or Scala to leverage the JVM's performance advantages over Python in Snowflake.

A data scientist uses bootstrapping to estimate the sampling distribution of a statistic calculated from a dataset stored in Snowflake. They observe that the bootstrap distribution is significantly different from the original data distribution. Which of the following statements best describes the possible reasons for this difference, considering both the theoretical underpinnings of bootstrapping and potential limitations?

• A. The original sample may not be representative of the population, and the bootstrap procedure is simply amplifying the biases present in the original sample. Additionally, the statistic itself may be highly sensitive to outliers or specific data points, leading to a distorted bootstrap distribution.
• B. The difference is unexpected; the bootstrap distribution should always closely resemble the original data distribution, regardless of the statistic being estimated.
• C. Bootstrapping always provides accurate estimates of sampling distributions, any significant difference indicates an error in the code implementation.
• D. The statistic being estimated is inherently unstable and has a high variance, causing the bootstrap distribution to be wider and potentially different in shape compared to the original data distribution. This is a normal outcome when dealing with such statistics.
• E. Bootstrapping is only appropriate for normally distributed data; if the original data is not normal, the bootstrap distribution will inevitably differ significantly.